The proposed project seeks to initiate in a major urban community a long-term prospective study of: (1) the impact on health (in conjunction wit other psychosocial risk factors) of psychological attributes (i.e., cognitive beliefs and abilities, affective states or traits, secular and religious world views, etc.); (2) their role in mediating and moderating the health effects of the socioeconomic and racial/ethnic statuses of both individuals and the social contexts/environments in which they live; and (3) the role of the sociodemograhic characteristics of individuals and their neighborhoods/social contexts in shaping health-related psychological attributes, in conjunction with other psychosocial factors (e.g., stress, social relationships and supports, etc.). The proposed research has several unique and distinguishing features. First, it will study for the first time in a representative community sample, the discriminant and convergent validity of a broad array of psychological attributes in relation to each other and to both their health sequelae and their sociodemographic and psychosocial antecedents. Second, it will do this in a thoroughly multilevel design, embedding a new survey of a major urban population in the context of an ongoing study which has been making the most thorough assessment ever of the social environment in which a large urban population resides. Finally, it will assess both self-reported and biomedical health outcomes and pathways to those outcomes. All of these data will be analyzed via the latest methods of multilevel analysis (often termed hierarchical linear modeling). In the proposed grant period, the project will design, execute, and thoroughly document and analyze a face-to-face survey of a socioeconomically and racial/ethnically diverse probability sample of 3,304 adults aged 25 and over in a major American city, integrating the survey data with extensive and intensive assessments of the residential social contexts in which respondents reside, and with anthropometric (e.g., body mass and waist-hip ratios) and hemodynamic (e.g., blood pressure data collected from all willing respondents during their interview, and serologic and neuroendocrine data from blood and saliva samples on almost 25% of the sample (for whom we will also have social contextual data enriched by systemic social observations). The study will importantly advance the understanding of socioeconomic and racial/ethnic disparities in health, a major priority of the Public Health Service and National Institutes of Health for the new millenium.